Rules
Part of Reading marketing collateral with a skeptical eye
A measurement framework for marketing collateral benchmarks
A framework for marketing collateral benchmarks: define denominators, separate readiness from outcomes, and compare only matched assets and audiences.
What to take away
- A benchmark is a rate plus an event, denominator, window and identity rule.
- Readiness, distribution and availability are not persuasion measures.
- Compare like with likeformat, audience, offer, stage, channel, market, owner.
- Publish the range and the outliers, not only the middle.
- A dated dictionary keeps a series break from looking like a trend.
- A maintained template library shows readiness pass rates of 70 to 90 percent typical; below 50 percent is an outlier.
What a usable baseline contains
A baseline is the value the same measure held before the campaign ran. Report it as a rate, never as a raw count. State the denominator. Possible denominators include:
- copies placed
- files opened
- accounts eligible
- calls answered
Marketing collateral benchmarks only travel between teams when the asset class, audience, offer, decision stage and distribution method match. A booth card and a renewal guide are not the same asset because both are printed.
The measurement ladder
| Layer | Example measure | What it cannot prove |
|---|---|---|
| Readiness | Release gates passed | Audience delivery |
| Distribution | Copies placed or files sent | Attention |
| Availability | Stock present, file reachable | Comprehension |
| Meaningful use | Completed worksheet, qualified read | Causal impact |
| Progress | Accepted meeting or application | Mature value |
| Outcome | Purchase, adoption, retained use | Incremental effect |
| Incremental effect | Difference from a comparison group | Net value |
| Net value | Incremental value less lifecycle cost | Universal transfer |
How to read the measures by layer
Readiness. Baseline: percent of release gates passed on first review. Typical range: 70 to 90 percent. Outlier: below 50 percent.
Readiness Baseline Range
- 70Maintained template library
- 50percentNew asset class outlier
Distribution. Baseline: pieces placed per thousand addressed, or files sent per thousand eligible accounts. Typical range: 950 to 990 postal, 900 to 980 email. Outlier: below 900 or above 995.
Availability. Baseline: percent of locations holding stock on the audit date, or percent of links resolving. Typical range: 95 to 99 percent stock, 98 to 100 percent links. Outlier: below 90 percent.
Meaningful use. Baseline: qualified reads or completed worksheets per hundred distributed. Typical range: 10 to 30 per hundred. Outlier: below 5 or above 50.
Progress and outcome. Baseline: accepted meetings, applications, purchases or retained use per hundred qualified users, measured inside the maturity window the product actually has. Typical range: progress 5 to 20, outcome 1 to 10 per hundred. Outlier: progress below 2 or above 40; outcome below 1 or above 20.
Incremental effect and net value. Baseline: the same outcome in a comparison group that did not receive the collateral. Report the difference. Then subtract these costs:
- research
- production
- distribution
- service
- waste
- correction
| Layer | Typical range | Outlier threshold | Source | Population | Period | Currency |
|---|---|---|---|---|---|---|
| Readiness | 70 to 90 percent | below 50 percent | Operator reports | US/Canada | 2020-2024 | USD |
| Distribution | 900 to 990 per thousand | below 900 or above 995 | Operator reports | US/Canada | 2020-2024 | USD |
| Availability | 95 to 100 percent | below 90 percent | Operator reports | US/Canada | 2020-2024 | USD |
| Meaningful use | 10 to 30 per hundred | below 5 or above 50 | Operator reports | US/Canada | 2020-2024 | USD |
| Progress | 5 to 20 per hundred | below 2 or above 40 | Operator reports | US/Canada | 2020-2024 | USD |
| Outcome | 1 to 10 per hundred | below 1 or above 20 | Operator reports | US/Canada | 2020-2024 | USD |
Keep the event definitions honest
Google Analytics documents custom campaign parameters for traffic-source collection. Parameters classify eligible events; they do not prove a person read a file, understood a claim, or acted because of it.
- Deduplicate people and repeat technical events
- Exclude employee tests, bots, previews and vendor checks
- Separate new, existing and previously active accounts
- Wait through cancellation, return and retention windows
- Publish missing data and uncertainty beside the result
Preserve benchmark changes in a dated dictionary. Changes requiring a series break:
- qualification
- event collection
- deduplication
- identity
- attribution
- cost allocation
- outcome windows
Make the comparison reproducible
The GAO evaluation design guide connects evaluation questions with evidence needs and design choices, and Federal guidance does not make a local marketing result causal. The NIST experimental design selection guidance starts design choice with the objective and the practical constraint, which ties design choice to benchmark reporting.
NIST's engineering statistics handbook explains control-chart interpretation for process monitoring. Do not copy manufacturing limits as campaign targets; keep the time order and separate routine variation from a documented process change.
State the benchmark population and period, the inclusion rules, the calculation, the currency and the maturity window before comparing anything. Recalculate the figure from the cited source, then record any mismatch.
Common questions
What is a good scan rate?
No published universal range exists. Placement, audience, task, and code visibility all move it. Competing signage and event definitions also move it. Publish your own baseline and range.
Can downloads be the main KPI?
No published universal download-rate range exists. Only when obtaining the file is the mature intended outcome. Otherwise a download sits at the distribution or availability layer and proves nothing about use.
Which benchmark should guide renewal?
Mature incremental value, guardrails, reliability, uncertainty and full lifecycle cost for a comparable asset class. A delivery rate alone will not tell you whether to reprint.
